The 3D semantic labeling task involves predicting a semantic labeling of a 3D scan mesh.

Evaluation and metrics

Our evaluation ranks all methods according to the PASCAL VOC intersection-over-union metric (IoU). IoU = TP/(TP+FP+FN), where TP, FP, and FN are the numbers of true positive, false positive, and false negative pixels, respectively. Predicted labels are evaluated per-vertex over the respective 3D scan mesh; for 3D approaches that operate on other representations like grids or points, the predicted labels should be mapped onto the mesh vertices (e.g., one such example for grid to mesh vertices is provided in the evaluation helpers).



This table lists the benchmark results for the 3D semantic label scenario.


Method Infoavg ioubathtubbedbookshelfcabinetchaircountercurtaindeskdoorfloorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwallwindow
sorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
Mix3Dpermissive0.781 10.964 10.855 10.843 100.781 10.858 70.575 20.831 170.685 50.714 10.979 10.594 30.310 160.801 10.892 70.841 20.819 30.723 20.940 70.887 10.725 10
Alexey Nekrasov, Jonas Schult, Or Litany, Bastian Leibe, Francis Engelmann: Mix3D: Out-of-Context Data Augmentation for 3D Scenes. 3DV 2021 (Oral)
OccuSeg+Semantic0.764 20.758 420.796 160.839 110.746 80.907 10.562 30.850 120.680 70.672 50.978 20.610 10.335 80.777 40.819 290.847 10.830 10.691 70.972 10.885 20.727 8
O-CNNpermissive0.762 30.924 20.823 40.844 90.770 20.852 90.577 10.847 130.711 10.640 130.958 90.592 40.217 530.762 100.888 80.758 80.813 50.726 10.932 130.868 80.744 4
Peng-Shuai Wang, Yang Liu, Yu-Xiao Guo, Chun-Yu Sun, Xin Tong: O-CNN: Octree-based Convolutional Neural Networks for 3D Shape Analysis. SIGGRAPH 2017
DMF-Net0.752 40.906 40.793 190.802 250.689 220.825 250.556 40.867 80.681 60.602 250.960 70.555 140.365 30.779 30.859 150.747 100.795 170.717 30.917 160.856 150.764 2
PointTransformerV20.752 40.742 490.809 110.872 10.758 40.860 60.552 50.891 50.610 250.687 20.960 70.559 120.304 190.766 80.926 20.767 60.797 130.644 170.942 50.876 70.722 12
Xiaoyang Wu, Yixing Lao, Li Jiang, Xihui Liu, Hengshuang Zhao: Point Transformer V2: Grouped Vector Attention and Partition-based Pooling. NeurIPS 2022
PointConvFormer0.749 60.793 280.790 200.807 220.750 70.856 80.524 120.881 60.588 350.642 120.977 40.591 50.274 310.781 20.929 10.804 30.796 140.642 180.947 30.885 20.715 14
Wenxuan Wu, Qi Shan, Li Fuxin: PointConvFormer: Revenge of the Point-based Convolution.
BPNetcopyleft0.749 60.909 30.818 70.811 190.752 60.839 150.485 270.842 150.673 80.644 100.957 110.528 210.305 180.773 60.859 150.788 40.818 40.693 60.916 170.856 150.723 11
Wenbo Hu, Hengshuang Zhao, Li Jiang, Jiaya Jia, Tien-Tsin Wong: Bidirectional Projection Network for Cross Dimension Scene Understanding. CVPR 2021 (Oral)
MSP0.748 80.623 730.804 120.859 30.745 90.824 270.501 190.912 20.690 40.685 30.956 120.567 90.320 130.768 70.918 30.720 180.802 90.676 90.921 150.881 40.779 1
StratifiedFormerpermissive0.747 90.901 50.803 130.845 80.757 50.846 110.512 150.825 180.696 30.645 90.956 120.576 70.262 390.744 150.861 140.742 110.770 280.705 40.899 270.860 120.734 5
Xin Lai*, Jianhui Liu*, Li Jiang, Liwei Wang, Hengshuang Zhao, Shu Liu, Xiaojuan Qi, Jiaya Jia: Stratified Transformer for 3D Point Cloud Segmentation. CVPR 2022
Virtual MVFusion0.746 100.771 360.819 60.848 60.702 200.865 50.397 650.899 30.699 20.664 60.948 360.588 60.330 90.746 140.851 210.764 70.796 140.704 50.935 100.866 90.728 6
Abhijit Kundu, Xiaoqi Yin, Alireza Fathi, David Ross, Brian Brewington, Thomas Funkhouser, Caroline Pantofaru: Virtual Multi-view Fusion for 3D Semantic Segmentation. ECCV 2020
VMNetpermissive0.746 100.870 100.838 20.858 40.729 130.850 100.501 190.874 70.587 360.658 70.956 120.564 100.299 200.765 90.900 50.716 210.812 60.631 230.939 80.858 130.709 15
Zeyu HU, Xuyang Bai, Jiaxiang Shang, Runze Zhang, Jiayu Dong, Xin Wang, Guangyuan Sun, Hongbo Fu, Chiew-Lan Tai: VMNet: Voxel-Mesh Network for Geodesic-Aware 3D Semantic Segmentation. ICCV 2021 (Oral)
Retro-FPN0.744 120.842 170.800 140.767 370.740 100.836 180.541 70.914 10.672 90.626 140.958 90.552 150.272 320.777 40.886 100.696 280.801 100.674 100.941 60.858 130.717 13
EQ-Net0.743 130.620 740.799 150.849 50.730 120.822 290.493 250.897 40.664 100.681 40.955 160.562 110.378 10.760 110.903 40.738 120.801 100.673 110.907 200.877 50.745 3
Zetong Yang*, Li Jiang*, Yanan Sun, Bernt Schiele, Jiaya JIa: A Unified Query-based Paradigm for Point Cloud Understanding. CVPR 2022
MinkowskiNetpermissive0.736 140.859 130.818 70.832 120.709 170.840 140.521 140.853 110.660 120.643 110.951 260.544 160.286 260.731 160.893 60.675 350.772 260.683 80.874 460.852 180.727 8
C. Choy, J. Gwak, S. Savarese: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. CVPR 2019
IPCA0.731 150.890 60.837 30.864 20.726 140.873 20.530 110.824 190.489 680.647 80.978 20.609 20.336 70.624 320.733 420.758 80.776 240.570 480.949 20.877 50.728 6
PointTransformer++0.725 160.727 560.811 100.819 150.765 30.841 130.502 180.814 240.621 210.623 150.955 160.556 130.284 270.620 330.866 120.781 50.757 350.648 150.932 130.862 110.709 15
SparseConvNet0.725 160.647 700.821 50.846 70.721 150.869 30.533 90.754 380.603 310.614 180.955 160.572 80.325 110.710 170.870 110.724 160.823 20.628 240.934 110.865 100.683 21
MatchingNet0.724 180.812 250.812 90.810 200.735 110.834 190.495 240.860 100.572 420.602 250.954 190.512 240.280 280.757 120.845 240.725 150.780 220.606 330.937 90.851 190.700 18
INS-Conv-semantic0.717 190.751 450.759 330.812 180.704 190.868 40.537 80.842 150.609 270.608 210.953 210.534 170.293 220.616 340.864 130.719 200.793 180.640 190.933 120.845 230.663 26
PointMetaBase0.714 200.835 180.785 230.821 130.684 240.846 110.531 100.865 90.614 220.596 280.953 210.500 270.246 450.674 180.888 80.692 290.764 300.624 250.849 600.844 240.675 23
contrastBoundarypermissive0.705 210.769 390.775 270.809 210.687 230.820 320.439 520.812 250.661 110.591 300.945 450.515 230.171 710.633 290.856 170.720 180.796 140.668 120.889 350.847 210.689 20
Liyao Tang, Yibing Zhan, Zhe Chen, Baosheng Yu, Dacheng Tao: Contrastive Boundary Learning for Point Cloud Segmentation. CVPR2022
RFCR0.702 220.889 70.745 420.813 170.672 270.818 360.493 250.815 220.623 190.610 190.947 390.470 370.249 440.594 380.848 220.705 250.779 230.646 160.892 330.823 310.611 42
Jingyu Gong, Jiachen Xu, Xin Tan, Haichuan Song, Yanyun Qu, Yuan Xie, Lizhuang Ma: Omni-Supervised Point Cloud Segmentation via Gradual Receptive Field Component Reasoning. CVPR2021
One Thing One Click0.701 230.825 220.796 160.723 440.716 160.832 200.433 540.816 200.634 170.609 200.969 60.418 630.344 50.559 500.833 260.715 220.808 70.560 520.902 240.847 210.680 22
JSENetpermissive0.699 240.881 90.762 310.821 130.667 280.800 490.522 130.792 300.613 230.607 220.935 640.492 300.205 580.576 440.853 190.691 300.758 340.652 140.872 490.828 280.649 31
Zeyu HU, Mingmin Zhen, Xuyang BAI, Hongbo Fu, Chiew-lan Tai: JSENet: Joint Semantic Segmentation and Edge Detection Network for 3D Point Clouds. ECCV 2020
PicassoNet-IIpermissive0.696 250.704 600.790 200.787 290.709 170.837 160.459 370.815 220.543 520.615 170.956 120.529 190.250 420.551 550.790 340.703 260.799 120.619 280.908 190.848 200.700 18
Huan Lei, Naveed Akhtar, Mubarak Shah, and Ajmal Mian: Geometric feature learning for 3D meshes.
CU-Hybrid Net0.693 260.596 780.789 220.803 240.677 260.800 490.469 310.846 140.554 500.591 300.948 360.500 270.316 140.609 350.847 230.732 130.808 70.593 400.894 310.839 250.652 30
One-Thing-One-Click0.693 260.743 480.794 180.655 670.684 240.822 290.497 230.719 480.622 200.617 160.977 40.447 500.339 60.750 130.664 570.703 260.790 200.596 370.946 40.855 170.647 32
Zhengzhe Liu, Xiaojuan Qi, Chi-Wing Fu: One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation. CVPR 2021
Feature_GeometricNetpermissive0.690 280.884 80.754 370.795 280.647 330.818 360.422 560.802 280.612 240.604 230.945 450.462 410.189 660.563 490.853 190.726 140.765 290.632 220.904 220.821 340.606 46
Kangcheng Liu, Ben M. Chen: https://arxiv.org/abs/2012.09439. arXiv Preprint
FusionNet0.688 290.704 600.741 460.754 410.656 290.829 220.501 190.741 430.609 270.548 380.950 300.522 220.371 20.633 290.756 370.715 220.771 270.623 260.861 560.814 360.658 27
Feihu Zhang, Jin Fang, Benjamin Wah, Philip Torr: Deep FusionNet for Point Cloud Semantic Segmentation. ECCV 2020
Feature-Geometry Netpermissive0.685 300.866 110.748 390.819 150.645 350.794 530.450 420.802 280.587 360.604 230.945 450.464 400.201 610.554 520.840 250.723 170.732 440.602 350.907 200.822 330.603 49
KP-FCNN0.684 310.847 160.758 350.784 310.647 330.814 390.473 290.772 330.605 290.594 290.935 640.450 480.181 690.587 390.805 320.690 310.785 210.614 290.882 390.819 350.632 37
H. Thomas, C. Qi, J. Deschaud, B. Marcotegui, F. Goulette, L. Guibas.: KPConv: Flexible and Deformable Convolution for Point Clouds. ICCV 2019
VACNN++0.684 310.728 550.757 360.776 330.690 210.804 470.464 350.816 200.577 410.587 320.945 450.508 260.276 300.671 190.710 470.663 400.750 380.589 430.881 400.832 270.653 29
Superpoint Network0.683 330.851 150.728 510.800 270.653 310.806 450.468 320.804 260.572 420.602 250.946 420.453 470.239 480.519 610.822 270.689 330.762 320.595 390.895 300.827 290.630 38
PointContrast_LA_SEM0.683 330.757 430.784 240.786 300.639 370.824 270.408 600.775 320.604 300.541 400.934 680.532 180.269 350.552 530.777 350.645 500.793 180.640 190.913 180.824 300.671 24
VI-PointConv0.676 350.770 380.754 370.783 320.621 410.814 390.552 50.758 360.571 440.557 360.954 190.529 190.268 370.530 590.682 520.675 350.719 470.603 340.888 360.833 260.665 25
Xingyi Li, Wenxuan Wu, Xiaoli Z. Fern, Li Fuxin: The Devils in the Point Clouds: Studying the Robustness of Point Cloud Convolutions.
ROSMRF3D0.673 360.789 290.748 390.763 390.635 390.814 390.407 620.747 400.581 400.573 330.950 300.484 310.271 340.607 360.754 380.649 450.774 250.596 370.883 380.823 310.606 46
SALANet0.670 370.816 240.770 290.768 360.652 320.807 440.451 390.747 400.659 130.545 390.924 740.473 360.149 810.571 460.811 310.635 530.746 390.623 260.892 330.794 490.570 59
PointConvpermissive0.666 380.781 310.759 330.699 520.644 360.822 290.475 280.779 310.564 470.504 560.953 210.428 570.203 600.586 410.754 380.661 410.753 360.588 440.902 240.813 380.642 33
Wenxuan Wu, Zhongang Qi, Li Fuxin: PointConv: Deep Convolutional Networks on 3D Point Clouds. CVPR 2019
PointASNLpermissive0.666 380.703 620.781 250.751 430.655 300.830 210.471 300.769 340.474 710.537 420.951 260.475 350.279 290.635 270.698 510.675 350.751 370.553 570.816 680.806 400.703 17
Xu Yan, Chaoda Zheng, Zhen Li, Sheng Wang, Shuguang Cui: PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling. CVPR 2020
PPCNN++permissive0.663 400.746 460.708 540.722 450.638 380.820 320.451 390.566 750.599 330.541 400.950 300.510 250.313 150.648 240.819 290.616 590.682 620.590 420.869 520.810 390.656 28
Pyunghwan Ahn, Juyoung Yang, Eojindl Yi, Chanho Lee, Junmo Kim: Projection-based Point Convolution for Efficient Point Cloud Segmentation. IEEE Access
DCM-Net0.658 410.778 320.702 570.806 230.619 420.813 420.468 320.693 560.494 640.524 480.941 560.449 490.298 210.510 630.821 280.675 350.727 460.568 500.826 650.803 420.637 35
Jonas Schult*, Francis Engelmann*, Theodora Kontogianni, Bastian Leibe: DualConvMesh-Net: Joint Geodesic and Euclidean Convolutions on 3D Meshes. CVPR 2020 [Oral]
HPGCNN0.656 420.698 630.743 440.650 680.564 600.820 320.505 170.758 360.631 180.479 610.945 450.480 330.226 490.572 450.774 360.690 310.735 420.614 290.853 590.776 630.597 52
Jisheng Dang, Qingyong Hu, Yulan Guo, Jun Yang: HPGCNN.
SAFNet-segpermissive0.654 430.752 440.734 480.664 650.583 540.815 380.399 640.754 380.639 150.535 440.942 540.470 370.309 170.665 200.539 650.650 440.708 520.635 210.857 580.793 510.642 33
Linqing Zhao, Jiwen Lu, Jie Zhou: Similarity-Aware Fusion Network for 3D Semantic Segmentation. IROS 2021
RandLA-Netpermissive0.645 440.778 320.731 490.699 520.577 550.829 220.446 440.736 440.477 700.523 500.945 450.454 450.269 350.484 700.749 410.618 570.738 400.599 360.827 640.792 540.621 40
MVPNetpermissive0.641 450.831 190.715 520.671 620.590 500.781 590.394 660.679 590.642 140.553 370.937 610.462 410.256 400.649 230.406 780.626 540.691 590.666 130.877 420.792 540.608 45
Maximilian Jaritz, Jiayuan Gu, Hao Su: Multi-view PointNet for 3D Scene Understanding. GMDL Workshop, ICCV 2019
PointConv-SFPN0.641 450.776 340.703 560.721 460.557 630.826 240.451 390.672 610.563 480.483 600.943 530.425 600.162 760.644 250.726 430.659 420.709 510.572 470.875 440.786 580.559 64
PointMRNet0.640 470.717 590.701 580.692 550.576 560.801 480.467 340.716 490.563 480.459 660.953 210.429 560.169 730.581 420.854 180.605 600.710 490.550 580.894 310.793 510.575 57
FPConvpermissive0.639 480.785 300.760 320.713 500.603 450.798 510.392 670.534 800.603 310.524 480.948 360.457 430.250 420.538 570.723 450.598 640.696 570.614 290.872 490.799 430.567 61
Yiqun Lin, Zizheng Yan, Haibin Huang, Dong Du, Ligang Liu, Shuguang Cui, Xiaoguang Han: FPConv: Learning Local Flattening for Point Convolution. CVPR 2020
PD-Net0.638 490.797 270.769 300.641 730.590 500.820 320.461 360.537 790.637 160.536 430.947 390.388 700.206 570.656 210.668 550.647 480.732 440.585 450.868 530.793 510.473 82
PointSPNet0.637 500.734 520.692 650.714 490.576 560.797 520.446 440.743 420.598 340.437 710.942 540.403 660.150 800.626 310.800 330.649 450.697 560.557 550.846 610.777 620.563 62
SConv0.636 510.830 200.697 610.752 420.572 590.780 610.445 460.716 490.529 550.530 450.951 260.446 510.170 720.507 650.666 560.636 520.682 620.541 640.886 370.799 430.594 53
Supervoxel-CNN0.635 520.656 680.711 530.719 470.613 430.757 700.444 490.765 350.534 540.566 340.928 720.478 340.272 320.636 260.531 670.664 390.645 730.508 710.864 550.792 540.611 42
joint point-basedpermissive0.634 530.614 750.778 260.667 640.633 400.825 250.420 570.804 260.467 730.561 350.951 260.494 290.291 230.566 470.458 730.579 700.764 300.559 540.838 620.814 360.598 51
Hung-Yueh Chiang, Yen-Liang Lin, Yueh-Cheng Liu, Winston H. Hsu: A Unified Point-Based Framework for 3D Segmentation. 3DV 2019
MCCNNpermissive0.633 540.866 110.731 490.771 340.576 560.809 430.410 590.684 570.497 630.491 580.949 330.466 390.105 850.581 420.646 590.620 550.680 640.542 630.817 670.795 470.618 41
P. Hermosilla, T. Ritschel, P.P. Vazquez, A. Vinacua, T. Ropinski: Monte Carlo Convolution for Learning on Non-Uniformly Sampled Point Clouds. SIGGRAPH Asia 2018
PointMTL0.632 550.731 530.688 680.675 590.591 490.784 580.444 490.565 760.610 250.492 570.949 330.456 440.254 410.587 390.706 480.599 630.665 690.612 320.868 530.791 570.579 56
PointNet2-SFPN0.631 560.771 360.692 650.672 600.524 670.837 160.440 510.706 540.538 530.446 680.944 510.421 620.219 520.552 530.751 400.591 660.737 410.543 620.901 260.768 650.557 65
3DSM_DMMF0.631 560.626 720.745 420.801 260.607 440.751 710.506 160.729 470.565 460.491 580.866 880.434 520.197 640.595 370.630 600.709 240.705 540.560 520.875 440.740 730.491 77
APCF-Net0.631 560.742 490.687 700.672 600.557 630.792 560.408 600.665 620.545 510.508 530.952 250.428 570.186 670.634 280.702 490.620 550.706 530.555 560.873 470.798 450.581 55
Haojia, Lin: Adaptive Pyramid Context Fusion for Point Cloud Perception. GRSL
FusionAwareConv0.630 590.604 770.741 460.766 380.590 500.747 720.501 190.734 450.503 620.527 460.919 780.454 450.323 120.550 560.420 770.678 340.688 600.544 600.896 290.795 470.627 39
Jiazhao Zhang, Chenyang Zhu, Lintao Zheng, Kai Xu: Fusion-Aware Point Convolution for Online Semantic 3D Scene Segmentation. CVPR 2020
DenSeR0.628 600.800 260.625 800.719 470.545 650.806 450.445 460.597 700.448 770.519 510.938 600.481 320.328 100.489 690.499 720.657 430.759 330.592 410.881 400.797 460.634 36
SegGroup_sempermissive0.627 610.818 230.747 410.701 510.602 460.764 670.385 710.629 670.490 660.508 530.931 710.409 650.201 610.564 480.725 440.618 570.692 580.539 650.873 470.794 490.548 68
An Tao, Yueqi Duan, Yi Wei, Jiwen Lu, Jie Zhou: SegGroup: Seg-Level Supervision for 3D Instance and Semantic Segmentation. TIP 2022
SIConv0.625 620.830 200.694 630.757 400.563 610.772 650.448 430.647 650.520 570.509 520.949 330.431 550.191 650.496 670.614 610.647 480.672 670.535 670.876 430.783 590.571 58
HPEIN0.618 630.729 540.668 710.647 700.597 480.766 660.414 580.680 580.520 570.525 470.946 420.432 530.215 540.493 680.599 620.638 510.617 780.570 480.897 280.806 400.605 48
Li Jiang, Hengshuang Zhao, Shu Liu, Xiaoyong Shen, Chi-Wing Fu, Jiaya Jia: Hierarchical Point-Edge Interaction Network for Point Cloud Semantic Segmentation. ICCV 2019
SPH3D-GCNpermissive0.610 640.858 140.772 280.489 850.532 660.792 560.404 630.643 660.570 450.507 550.935 640.414 640.046 910.510 630.702 490.602 620.705 540.549 590.859 570.773 640.534 71
Huan Lei, Naveed Akhtar, and Ajmal Mian: Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds. TPAMI 2020
AttAN0.609 650.760 410.667 720.649 690.521 680.793 540.457 380.648 640.528 560.434 730.947 390.401 670.153 790.454 720.721 460.648 470.717 480.536 660.904 220.765 660.485 78
Gege Zhang, Qinghua Ma, Licheng Jiao, Fang Liu and Qigong Sun: AttAN: Attention Adversarial Networks for 3D Point Cloud Semantic Segmentation. IJCAI2020
wsss-transformer0.600 660.634 710.743 440.697 540.601 470.781 590.437 530.585 730.493 650.446 680.933 690.394 680.011 930.654 220.661 580.603 610.733 430.526 680.832 630.761 680.480 79
LAP-D0.594 670.720 570.692 650.637 740.456 770.773 640.391 690.730 460.587 360.445 700.940 580.381 710.288 240.434 750.453 750.591 660.649 710.581 460.777 720.749 720.610 44
DPC0.592 680.720 570.700 590.602 780.480 730.762 690.380 720.713 520.585 390.437 710.940 580.369 730.288 240.434 750.509 710.590 680.639 760.567 510.772 730.755 700.592 54
Francis Engelmann, Theodora Kontogianni, Bastian Leibe: Dilated Point Convolutions: On the Receptive Field Size of Point Convolutions on 3D Point Clouds. ICRA 2020
CCRFNet0.589 690.766 400.659 750.683 570.470 760.740 740.387 700.620 690.490 660.476 620.922 760.355 760.245 460.511 620.511 700.571 710.643 740.493 750.872 490.762 670.600 50
ROSMRF0.580 700.772 350.707 550.681 580.563 610.764 670.362 740.515 810.465 740.465 650.936 630.427 590.207 560.438 730.577 630.536 740.675 660.486 760.723 790.779 600.524 73
SD-DETR0.576 710.746 460.609 840.445 890.517 690.643 850.366 730.714 510.456 750.468 640.870 870.432 530.264 380.558 510.674 530.586 690.688 600.482 770.739 770.733 750.537 70
SQN_0.1%0.569 720.676 650.696 620.657 660.497 700.779 620.424 550.548 770.515 590.376 780.902 850.422 610.357 40.379 790.456 740.596 650.659 700.544 600.685 820.665 860.556 66
TextureNetpermissive0.566 730.672 670.664 730.671 620.494 710.719 750.445 460.678 600.411 830.396 760.935 640.356 750.225 500.412 770.535 660.565 720.636 770.464 790.794 710.680 830.568 60
Jingwei Huang, Haotian Zhang, Li Yi, Thomas Funkerhouser, Matthias Niessner, Leonidas Guibas: TextureNet: Consistent Local Parametrizations for Learning from High-Resolution Signals on Meshes. CVPR
DVVNet0.562 740.648 690.700 590.770 350.586 530.687 790.333 780.650 630.514 600.475 630.906 820.359 740.223 510.340 810.442 760.422 850.668 680.501 720.708 800.779 600.534 71
Pointnet++ & Featurepermissive0.557 750.735 510.661 740.686 560.491 720.744 730.392 670.539 780.451 760.375 790.946 420.376 720.205 580.403 780.356 810.553 730.643 740.497 730.824 660.756 690.515 74
GMLPs0.538 760.495 860.693 640.647 700.471 750.793 540.300 810.477 820.505 610.358 800.903 840.327 790.081 880.472 710.529 680.448 830.710 490.509 690.746 750.737 740.554 67
PanopticFusion-label0.529 770.491 870.688 680.604 770.386 820.632 860.225 910.705 550.434 800.293 860.815 890.348 770.241 470.499 660.669 540.507 760.649 710.442 850.796 700.602 890.561 63
Gaku Narita, Takashi Seno, Tomoya Ishikawa, Yohsuke Kaji: PanopticFusion: Online Volumetric Semantic Mapping at the Level of Stuff and Things. IROS 2019 (to appear)
subcloud_weak0.516 780.676 650.591 870.609 750.442 780.774 630.335 770.597 700.422 820.357 810.932 700.341 780.094 870.298 830.528 690.473 810.676 650.495 740.602 880.721 780.349 89
Online SegFusion0.515 790.607 760.644 780.579 800.434 790.630 870.353 750.628 680.440 780.410 740.762 920.307 810.167 740.520 600.403 790.516 750.565 810.447 830.678 830.701 800.514 75
Davide Menini, Suryansh Kumar, Martin R. Oswald, Erik Sandstroem, Cristian Sminchisescu, Luc van Gool: A Real-Time Learning Framework for Joint 3D Reconstruction and Semantic Segmentation. Robotics and Automation Letters Submission
3DMV, FTSDF0.501 800.558 820.608 850.424 910.478 740.690 780.246 870.586 720.468 720.450 670.911 800.394 680.160 770.438 730.212 880.432 840.541 860.475 780.742 760.727 760.477 80
PCNN0.498 810.559 810.644 780.560 820.420 810.711 770.229 890.414 830.436 790.352 820.941 560.324 800.155 780.238 880.387 800.493 770.529 870.509 690.813 690.751 710.504 76
3DMV0.484 820.484 880.538 890.643 720.424 800.606 900.310 790.574 740.433 810.378 770.796 900.301 820.214 550.537 580.208 890.472 820.507 900.413 880.693 810.602 890.539 69
Angela Dai, Matthias Niessner: 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation. ECCV'18
PointCNN with RGBpermissive0.458 830.577 800.611 830.356 930.321 900.715 760.299 830.376 870.328 900.319 840.944 510.285 840.164 750.216 910.229 860.484 790.545 850.456 810.755 740.709 790.475 81
Yangyan Li, Rui Bu, Mingchao Sun, Baoquan Chen: PointCNN. NeurIPS 2018
FCPNpermissive0.447 840.679 640.604 860.578 810.380 830.682 800.291 840.106 930.483 690.258 910.920 770.258 880.025 920.231 900.325 820.480 800.560 830.463 800.725 780.666 850.231 93
Dario Rethage, Johanna Wald, Jürgen Sturm, Nassir Navab, Federico Tombari: Fully-Convolutional Point Networks for Large-Scale Point Clouds. ECCV 2018
DGCNN_reproducecopyleft0.446 850.474 890.623 810.463 870.366 850.651 830.310 790.389 860.349 880.330 830.937 610.271 860.126 830.285 840.224 870.350 900.577 800.445 840.625 860.723 770.394 85
Yue Wang, Yongbin Sun, Ziwei Liu, Sanjay E. Sarma, Michael M. Bronstein, Justin M. Solomon: Dynamic Graph CNN for Learning on Point Clouds. TOG 2019
SurfaceConvPF0.442 860.505 850.622 820.380 920.342 880.654 820.227 900.397 850.367 860.276 880.924 740.240 890.198 630.359 800.262 840.366 870.581 790.435 860.640 850.668 840.398 84
Hao Pan, Shilin Liu, Yang Liu, Xin Tong: Convolutional Neural Networks on 3D Surfaces Using Parallel Frames.
PNET20.442 860.548 830.548 880.597 790.363 860.628 880.300 810.292 880.374 850.307 850.881 860.268 870.186 670.238 880.204 900.407 860.506 910.449 820.667 840.620 880.462 83
Tangent Convolutionspermissive0.438 880.437 910.646 770.474 860.369 840.645 840.353 750.258 900.282 920.279 870.918 790.298 830.147 820.283 850.294 830.487 780.562 820.427 870.619 870.633 870.352 88
Maxim Tatarchenko, Jaesik Park, Vladlen Koltun, Qian-Yi Zhou: Tangent convolutions for dense prediction in 3d. CVPR 2018
3DWSSS0.425 890.525 840.647 760.522 830.324 890.488 930.077 940.712 530.353 870.401 750.636 940.281 850.176 700.340 810.565 640.175 940.551 840.398 890.370 940.602 890.361 87
SPLAT Netcopyleft0.393 900.472 900.511 900.606 760.311 910.656 810.245 880.405 840.328 900.197 920.927 730.227 910.000 950.001 950.249 850.271 930.510 880.383 910.593 890.699 810.267 91
Hang Su, Varun Jampani, Deqing Sun, Subhransu Maji, Evangelos Kalogerakis, Ming-Hsuan Yang, Jan Kautz: SPLATNet: Sparse Lattice Networks for Point Cloud Processing. CVPR 2018
ScanNet+FTSDF0.383 910.297 930.491 910.432 900.358 870.612 890.274 850.116 920.411 830.265 890.904 830.229 900.079 890.250 860.185 910.320 910.510 880.385 900.548 900.597 920.394 85
PointNet++permissive0.339 920.584 790.478 920.458 880.256 930.360 940.250 860.247 910.278 930.261 900.677 930.183 920.117 840.212 920.145 930.364 880.346 940.232 940.548 900.523 930.252 92
Charles R. Qi, Li Yi, Hao Su, Leonidas J. Guibas: pointnet++: deep hierarchical feature learning on point sets in a metric space.
SSC-UNetpermissive0.308 930.353 920.290 940.278 940.166 940.553 910.169 930.286 890.147 940.148 940.908 810.182 930.064 900.023 940.018 950.354 890.363 920.345 920.546 920.685 820.278 90
ScanNetpermissive0.306 940.203 940.366 930.501 840.311 910.524 920.211 920.002 950.342 890.189 930.786 910.145 940.102 860.245 870.152 920.318 920.348 930.300 930.460 930.437 940.182 94
Angela Dai, Angel X. Chang, Manolis Savva, Maciej Halber, Thomas Funkhouser, Matthias Nießner: ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes. CVPR'17
ERROR0.054 950.000 950.041 950.172 950.030 950.062 950.001 950.035 940.004 950.051 950.143 950.019 950.003 940.041 930.050 940.003 950.054 950.018 950.005 950.264 950.082 95